EP2756361A1 - Commande d'une machine - Google Patents

Commande d'une machine

Info

Publication number
EP2756361A1
EP2756361A1 EP11779378.6A EP11779378A EP2756361A1 EP 2756361 A1 EP2756361 A1 EP 2756361A1 EP 11779378 A EP11779378 A EP 11779378A EP 2756361 A1 EP2756361 A1 EP 2756361A1
Authority
EP
European Patent Office
Prior art keywords
determined
observations
relaxed
subset
assumptions
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
EP11779378.6A
Other languages
German (de)
English (en)
Inventor
Stephan Grimm
Thomas Hubauer
Mikhail Roshchin
Holger Stender
Stuart Watson
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Siemens AG
Original Assignee
Siemens AG
Siemens Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Siemens AG, Siemens Corp filed Critical Siemens AG
Publication of EP2756361A1 publication Critical patent/EP2756361A1/fr
Withdrawn legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
    • G05B23/0275—Fault isolation and identification, e.g. classify fault; estimate cause or root of failure
    • G05B23/0278—Qualitative, e.g. if-then rules; Fuzzy logic; Lookup tables; Symptomatic search; FMEA

Definitions

  • the invention relates to a method and a device for controlling a machine, in particular a gas or steam turbine.
  • the data obtained from the sensors are processed, analyzed and interpreted.
  • Model-based information interpretation (and its application as part of a model-based diagnosis) is becoming increasingly important.
  • Model-based methods have the advantage of an explicit and comprehensible description of the domain (eg of the technical system to be diagnosed).
  • Such an explicit model can be investigated and ver ⁇ stood, which promotes the acceptance of the user especially with regard to a diagnosis or interpretation resulting ⁇ nis.
  • the models can be adapted to new machines with a manageable effort, expanded with new domain knowledge, and even checked for correctness depending on the type of display.
  • a vocabulary of the model for a ne human-machine interaction and thus be used for the realization of an interactive interpretation process.
  • the task is for a given Mo ⁇ dell ⁇ (also referred to as a "theory") and a lot of Be ⁇ observations ⁇ an amount A of assumptions (usually as a subset .4 CA from all sorts of assumptions -4) such to determine that the observations ö are explained by the model T as well as the assumptions .4 CA.
  • the problem is formulated as an optimization problem, ie we search for the "best" such set of assumptions (depending on the optimality criterion, eg the smallest quantity or the quantity with the lowest weight).
  • the object of the invention is to avoid the above-ge ⁇ called disadvantages and to provide a way to ⁇ From production in relation to incorrect observations. This object is achieved according to the features of the independent claims. Preferred embodiments are in particular removable from the dependent claims.
  • the present approach allows for interactive diagnosis of different data, with the diagnosis based on, for example, a model of the turbine.
  • missing observations and / or faulty models can at least partially be compensated.
  • Thus can be herlite before ⁇ before they enter potential errors.
  • it is possible that the function of the technical system is ensured despite faulty models.
  • a diagnostic model which includes observations (eg observable features) of parts of the turbine, which parts may have an effect on the diagnosis of the entire turbine.
  • the model can be created from the beginning or it can be based models, existing (design) data or similar. be considered for the creation of the model.
  • the information or data, which are taken into account when creating the model can, for example, be present in UML (Unified Modeling Language) or another domain-specific language or be noted down.
  • representation or description languages it is possible to detect relational structures and make it accessible to an automatic evaluation.
  • a particular advantage is that an incomplete and possibly also faulty model can be used.
  • erroneous and incomplete information can lead to a specific diagnosis.
  • the implementation can be done in different description languages.
  • control can relate to or include control, diagnosis or other processing of data of the technical system.
  • control also includes a diagnosis, such as the use of information during a maintenance interval.
  • the machine comprises a turbine ⁇ , in particular a gas turbine or a steam turbine.
  • the optimality Optimriossbeg- different reefs may be used, such as minimum and / or maximum number of elements subset and / or superset relationship, or minimum and / or maximum sum of the weights of the ent ⁇ held elements.
  • the order ⁇ is based on the orders ⁇ A and o as follows:
  • the relaxed abduction problem is solved by transforming the relaxed abduction problem into a hypergraph so that tuples (A, ()) are encoded by pareto-optimal paths in the hypergraph. It is also a development that the Pareto-optimal paths are determined by means of a label approach.
  • hyper-edges of the hyp graph are induced by rewriting prescribed rules.
  • V ⁇ (ACB), (AC 3r.B) ⁇ A, B e NJ, re N R ⁇ ,
  • V T ⁇ (ACA), (ACT) IA e JV ⁇ C 1
  • a next embodiment is that shortest hyper paths are determined taking into account two preferences.
  • the shortest hyperpaths are determined taking into account two preferences by means of a label correction algorithm.
  • a further development consists in the fact that the labels code pareto-optimal paths to the hitherto found nodes of the hypergraphene.
  • An additional embodiment is that changes along the hyper-edges are propagated by means of a meet operator and / or by means of a join operator.
  • Another embodiment is that the relaxed abduction problem is determined by means of a description logic.
  • the processing unit may be a processor unit and / or an at least partially hardwired or logic circuit arrangement which is set up, for example, such that the method can be carried out as described herein.
  • Said processing unit can be any type of processor or computer or computer with correspondingly necessary peripherals (memory, input / output
  • the above explanations regarding the method apply to the device accordingly.
  • the device may be implemented in one component or distributed in several components.
  • angebun ⁇ eg, the Internet
  • Computer network proposed comprising at least one of the devices described herein.
  • the solution presented herein further includes a computer program product directly loadable into a memory of a digital computer comprising program code portions adapted to perform steps of the method described herein.
  • a computer-readable storage medium e.g. any memory comprising computer-executable instructions (e.g., in the form of program code) adapted for the computer to perform steps of the method described herein.
  • Fig.l is a schematic representation of an algorithm in
  • FIG. 3 shows a schematic block diagram with control units for controlling a technical system.
  • the solution proposed here includes the fol ⁇ constricting steps:
  • the definition of logic-based abduction is formally relaxed, the more important characteristics defined problems (such as the provability of statements about correctness and existence of solutions, etc.) ⁇ th preserver.
  • the relaxed abduction problem is translated into a hypergraph such that optimal pairs (A, 0) are complemented by pareto-optimal paths in the induced hypergraph. be diert.
  • a label approach is used.
  • model-based information interpretation (and thus model-based diagnostics) is significantly extended by the approach proposed here, since now also situations with an abundance of observation data (or a poorly formulated model) can be processed.
  • the solution shown here is kon ⁇ conservatively, that is, in cases where a conventional method provides a solution to a corresponding solution is also provided by the approach proposed here.
  • ⁇ can be understood as an inversion of the mode-ponens rule, which makes it possible to derive ⁇ as a hypothetical explanation for the occurrence of ⁇ , on condition that the presence of ⁇ in a sense justifies ⁇ .
  • This general formulation can not assume causality between ⁇ and ⁇ .
  • Different understandings of how ⁇ justifies the presence of ⁇ lead to different understandings of abductive interference, such as a so-called set-cover-based approach, logic-based approaches, or a knowledge-level approach.
  • Ai is "not worse” than A j , being an indifference and a strict preference
  • Ai Aj ⁇ A j ⁇ Ai with Ai A j be determined. Then a (normal) preference-based abduction problem can be defined as follows:
  • Definition 1 preference-based abduction problem
  • Typical preference orders for quantities are or include
  • weight-based systems which are defined by a function on w ⁇ which subsets of A nume i- see assigns weights:
  • the first two orders of preference prefer a set A over any of their subsets; this monotonicity is formalized in Definition 2 below.
  • An order - ⁇ ⁇ ) over sets is monotone (strictly monotone) for a subset relationship, if S 'S implies S' S (or S f CS implies S 'S). Conversely, an order (- ⁇ ) is anti-monotone (strictly anti-monotonic) for a subset relationship if 5 " ⁇ > S implies S 'S (S f D 5 * implies S' - ⁇ S).
  • a production system comprising a diagnostic unit, where ⁇ in the production system using a model was tillbil ⁇ det.
  • the model indicates that a vary ⁇ de-energizes temporary failure of a main control unit expresses while the communication links remain functional and mechanical grippers of the production system is not affected (the Beobachtun ⁇ gen are as causal consequence of diagnostic models).
  • Order is based on the orders ⁇ . and o as follows:
  • the (NF1) describes a concept inclusion "all objects of a class A are also objects of a class B".
  • the (NF2) describes: "if an object belongs to the class A ⁇ and to the class A, then it also belongs to the class B". This can be abbreviated as "A ⁇ and A implies B”.
  • the (NF3) denotes: "if an object to the class A ⁇ heard, then it is connected with at least one object of class B over ei ⁇ ne relation r".
  • the (NF4) describes: "If an object is connected to at least one object of class A by means of a relation r, then this object belongs to the class B".
  • the normal forms (NF5) and (NF6) are given correspondingly for the roles r ll ⁇ 2, s € N R.
  • a structure produced according to the rules of a graph made light ⁇ a derivation of subsumptions.
  • the axiom oriented representation allows a high flexibil ⁇ ty and reuse of information.
  • any normalized axioms can be mapped as a hypergraph (or as a corresponding representation of such a hypergraph) whose nodes have axioms of type (NF1) and (NF3) over the concepts and the Role names used in the Axiommenge are (corresponding to all as a premise or Conclusion in a derivation step admitted statements).
  • Hyper-edges of the hypergraph are induced by rewriting the rules (CR1) to (CR6); for example, induces an instantiation of the rule (CR4), the CCF of rD C and D ⁇ E by means of the axiom - ⁇ rF CF derives a Hype kan ⁇ te
  • V ⁇ (ACB), (AC 3r, B) ⁇ A, B e iVj, re N R ⁇
  • V T ⁇ (ACA), (AT) ⁇ AJ ⁇ C T ⁇ CV the set of final states and E the set of all hyperkants
  • T (e) CV on the basis of one of the rules (CR1) to (CR6) right ⁇ manufactures.
  • the edge weight w (e) is defined by
  • Hyperpath there is a hyper-path from X to t, if there is a hyper-edge connecting a given set of nodes Y to t, and each t ⁇ € .Y can be reached from X via a hyper-path.
  • Px , Vi are hyperpaths from X to,;:
  • VDV X , t ⁇ t ⁇ UU Wi6 r (e) V X # i '
  • the meet operator is defined as follows:
  • the join operator can be defined as follows:
  • FIG. 1 is a schematic representation of an algorithm in pseudo-code notation for exemplifying the propagation of the labels based on rule (CR4).
  • the labels for the hyperpath of the relaxed reduction problem are generated by means of the algorithm shown in FIG. In lines 1 to 4 an initialization takes place and in the following lines of the code fragment shown the labels are assigned or changes of the labels are propagated.
  • Hyper paths in U RA I begin at r, make Ablei ⁇ obligations are. Label that will be constructed based on this hyper paths can be used to encode relevant information that will be used during this derivation. According to claim 2, it suffices to show that the proposed algorithm correctly determines the labels of all pareto-optimal paths in U KA I starting at T.
  • the worst-case size of nodes Labels least reduced to a factor ⁇ , c)
  • the expected number of non-dominated paths to a state can also be determined as follows: Any two orders are taken over elements of ⁇ and ö such that any subset directly as a binary vector of length
  • weights for observations and / or abductible axioms can be given so that only those solutions are dropped that are significantly worse than others.
  • weights can be used to calculate limits for a maximum score that can be achieved by a partial solution; this score can be used as a pruning criterion.
  • the present approach provides a possibility of re ⁇ laxed abduction for a description logic.
  • the relaxed abduction extends the logic-based abduction by the possibility of interpreting false information regarding incomplete models.
  • a solution of relaxed abduction over EC + knowledge bases based on pareto-optimal hyperpathias in the derivation graph is presented. This approach also has significant advantages in terms of performance over a mere enumeration despite the inherent exponential growth of node labels.
  • the proposed algorithm is accordingly applicable to other description logics for which a subsumption can be decided by completion. This is e.g. in the description logic of the case.
  • a relaxed abduction problem for the technical system is determined, e.g. based on data from transducers or sensors or other collectable data concerning the technical system.
  • the relaxed abduction problem is solved.
  • the technical system may be a technical facility, a manufacturing facility, a process control, a power plant or the like.
  • control unit 301 which is arranged by way of example within a technical An ⁇ position 302nd
  • a control unit 303 is provided, which is arranged separately from the technical system 302 and connected to it via a network 304, for example the Internet.
  • Both control units 301, 303 can be used to control the technical system 302; in particular, at least one of the control units 301, 303 can carry out a diagnosis for the technical system 302 and / or set parameters of the technical system 302.
  • the technical system is a machine that particular ⁇ least comprises at least one gas turbine and / or Minim a steam turbine.
  • the technical sys- tem may include several turbines and optionally be provided centrally for the Diag ⁇ nose several such turbines.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Automation & Control Theory (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Software Systems (AREA)
  • Medical Informatics (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Quality & Reliability (AREA)
  • Machine Translation (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

L'invention vise à permettre une abduction efficace, y compris pour des observations inexactes ou insuffisamment modélisées. A cet effet, il est prévu un problème d'abduction assoupli pour expliquer la plus grande part possible des observations avec le moins d'hypothèses possible. Cette mesure permet de déterminer, sur la base de deux ordres de préférence, des tuples concernant une quantité partielle des observations et une quantité partielle des hypothèses, de sorte que la théorie, conjointement avec la quantité partielle des hypothèses, explique la quantité partielle des observations. La formulation sous forme de problème d'optimisation à critères multiples permet de ne plus devoir opposer les hypothèses rencontrées aux observations expliquées. Des propriétés déterminées de la quantité de résultats (comme l'exactitude, l'intégralité, etc.) sont vérifiées sur la base du fondement formel de la mise en équation, ce qui est particulièrement avantageux pour des applications critiques en termes de sécurité. Il est possible d'influer sur la complexité du processus de résolution de problème sur la base du choix de la langue de représentation concernée et des relations de préférence et de l'adapter de manière flexible pour ce qui est des demandes de domaines. Ladite invention peut s'appliquer à n'importe quelle machine, par ex. des turbines à gaz ou des turbines à vapeur.
EP11779378.6A 2011-10-28 2011-10-28 Commande d'une machine Withdrawn EP2756361A1 (fr)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/EP2011/069048 WO2013060389A1 (fr) 2011-10-28 2011-10-28 Commande d'une machine

Publications (1)

Publication Number Publication Date
EP2756361A1 true EP2756361A1 (fr) 2014-07-23

Family

ID=44910198

Family Applications (1)

Application Number Title Priority Date Filing Date
EP11779378.6A Withdrawn EP2756361A1 (fr) 2011-10-28 2011-10-28 Commande d'une machine

Country Status (3)

Country Link
US (1) US9897983B2 (fr)
EP (1) EP2756361A1 (fr)
WO (1) WO2013060389A1 (fr)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102011079034A1 (de) 2011-07-12 2013-01-17 Siemens Aktiengesellschaft Ansteuerung eines technischen Systems
CN111898372B (zh) * 2020-07-10 2022-08-16 中国标准化研究院 识别设计理性的方法、装置及计算机存储介质

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1994009427A1 (fr) * 1992-10-13 1994-04-28 Hewlett-Packard Company Procede et appareil de gestion de systeme
US6012152A (en) * 1996-11-27 2000-01-04 Telefonaktiebolaget Lm Ericsson (Publ) Software fault management system
US20070288405A1 (en) 2006-06-07 2007-12-13 Motorola, Inc. Problem solving mechanism selection facilitation apparatus and method
US7720787B2 (en) * 2006-06-10 2010-05-18 Distip Pty Limited Intelligence analysis method and system using subjective logic
US8014291B2 (en) * 2006-11-28 2011-09-06 Cisco Technology, Inc. Relaxed constrained shortest path first (R-CSPF)
US7962321B2 (en) * 2007-07-10 2011-06-14 Palo Alto Research Center Incorporated Modeling when connections are the problem
US8001423B2 (en) 2008-09-26 2011-08-16 Bae Systems Information And Electronic Systems Integration Inc. Prognostic diagnostic capability tracking system
US20130054506A1 (en) 2010-04-23 2013-02-28 Siemens Aktiengesellschaft Method and device for controlling an industrial system
EP2753994A1 (fr) * 2011-10-28 2014-07-16 Siemens Aktiengesellschaft Traitement d'un système technique

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
None *
See also references of WO2013060389A1 *

Also Published As

Publication number Publication date
US9897983B2 (en) 2018-02-20
US20140277755A1 (en) 2014-09-18
WO2013060389A1 (fr) 2013-05-02

Similar Documents

Publication Publication Date Title
DE602005004997T2 (de) Diagnostisches Verfahren und System
DE69523588T2 (de) Gerät und methode zur ereigniskorrelation und problemmeldung
DE69113077T2 (de) Modellgestütztes deduktives System für Netzwerksfehlerdiagnose.
DE69223787T2 (de) System fuer qualitative schlussfolgerung mit paralleler verarbeitung
DE102012207437B3 (de) Verfahren zum Betrieb einer speicherprogrammierbaren Steuerung
DE3856379T2 (de) Expertensystem mit verfahrenssteuerung
DE102004024262A1 (de) Wissensbasiertes Diagnosesystem für ein komplexes technisches System mit zwei getrennten Wissensbasen zur Verarbeitung technischer Systemdaten und zur Verarbeitung von Kundenbeanstandungen
DE68929289T2 (de) Expertensystem für fehlerdiagnose
DE60212121T2 (de) Erzeugung von prozessverwandten daten
DE69712119T2 (de) Ein effizientes datengesteuertes Theorierevisionssystem
EP0768584B1 (fr) Méthode de surveillance d'une machine ou installation
EP1020815A2 (fr) Outil et méthode pour diagnose automatique d'un système technique avec réemploi efficace d'information
DE102019217613A1 (de) Verfahren zur diagnose eines motorzustands und diagnostisches modellierungsverfahren dafür
WO2006133865A1 (fr) Priorisation dynamique d'operations de verification dans un diagnostic d'atelier
AT523586A1 (de) Verfahren zur zumindest teilweise dezentralen Berechnung des Gesundheitszustandes von mindestens einer Windkraftanlage
EP3781922A1 (fr) Système de diagnostic et procédé de traitement de données d'un véhicule à moteur
WO2001059535A1 (fr) Procede pour l'exploitation d'une installation technique
DE102020211679A1 (de) Computer-implementiertes system und verfahren mit einem digitalen zwilling und einer graphen-basierten struktur
EP3968149A1 (fr) Génération des règles de commande à partir d'un régulateur schématisé
WO2013060389A1 (fr) Commande d'une machine
EP2712429A1 (fr) Commande d'un système technique
Cacciabue et al. A cognitive model in a blackboard architecture: synergism of AI and psychology
DE112019007691T5 (de) System zum planen von roboterhandlungen, robotersystem, roboteraufgabenverifizierungssystem und verfahren zum planen von roboterhandlungen
DE69033526T2 (de) Industrielles Expertensystem
DE69712117T2 (de) Delta-Lernsystem zum Benutzen von Expertenberatung zur Revision von Fehlerhierarchien in diagnostischen Expertensystemen

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20140414

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

DAX Request for extension of the european patent (deleted)
RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: SIEMENS AKTIENGESELLSCHAFT

17Q First examination report despatched

Effective date: 20171109

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN

18D Application deemed to be withdrawn

Effective date: 20180320